Skip to main content

Scheduling

The Frank!Framework uses the Quartz scheduler for periodic adapter execution.

Configuration​

Schedules are defined in the Frank configuration XML:

<MSGScheduler>
    <MSGSchedulerJob name="MyJob" cronExpression="0 0/5 * * * ?" adapterName="MyAdapter"/>
</MSGScheduler>

Schedules can also be uploaded via the Frank!Console.

Cron Expressions​

Standard Quartz cron format with six or seven fields:

seconds minutes hours day-of-month month day-of-week [year]

Examples:

ExpressionMeaning
0 0/5 * * * ?Every 5 minutes
0 0 8 * * ?Daily at 08:00
0 0 0 1 * ?First day of each month at midnight
0 30 6 ? * MON-FRIWeekdays at 06:30
0 0 22 ? * SAT,SUNSaturday and Sunday at 22:00

Console​

The Scheduling page in the Frank!Console shows:

  • Job groups with their triggers
  • Next fire time for each trigger
  • Previous fire time
  • Trigger state (NORMAL, PAUSED)
  • Pause and resume controls

Pause and Resume​

To temporarily stop a scheduled job without removing it:

  1. Navigate to Scheduling in the Frank!Console
  2. Find the job in the list
  3. Click Pause next to the trigger
  4. The trigger state changes to PAUSED — no further executions occur
  5. Click Resume when ready to restart the schedule

Pausing is useful during maintenance windows or when a downstream dependency is unavailable.

Pause All / Resume All​

The page header provides buttons to pause or resume all triggers at once. This is equivalent to a "maintenance mode" for all scheduled processing.

Uploading a Schedule​

Schedules can be defined outside the Frank configuration XML and uploaded via the console:

  1. Navigate to Scheduling
  2. Click Upload
  3. Provide:
    • Job name — unique identifier
    • Cron expression — the schedule
    • Adapter name — the adapter to invoke
    • Configuration name — which configuration contains the adapter
    • Message (optional) — input message to pass to the adapter
  4. Click Save

Uploaded schedules are stored in the database and survive application restarts (if using a database-backed job store).

Modifying an Uploaded Schedule​

To change a schedule:

  1. Delete the existing job from the Scheduling page
  2. Upload a new job with the updated cron expression

There is no in-place edit; delete and recreate is the workflow.

Persistence​

When configured with a database-backed job store, schedules survive application restarts. Without database persistence, schedules are recreated from the configuration on startup.

Configure a database-backed job store by pointing it at the application datasource:

quartz.scheduler.jobStore.class=org.quartz.impl.jdbcjobstore.JobStoreTX
quartz.scheduler.jobStore.dataSource=quartzDS
quartz.scheduler.dataSource.quartzDS.jndiURL=java:comp/env/jdbc/${instance.name.lc}

The jndiURL must match the JNDI name of the datasource configured in the application server or Spring context.

Interval-Based Scheduling​

For simple interval scheduling (no cron complexity), use MSGSchedulerJob with an interval attribute inside <MSGScheduler>:

<MSGScheduler>
    <MSGSchedulerJob name="PollJob" interval="300000" adapterName="PollAdapter"/>
</MSGScheduler>

The interval is in milliseconds (300000 = 5 minutes).